Protection configuration-oriented short-circuit parameter rapid evaluation method, system, equipment and medium

Abstract

The invention discloses a protection configuration-oriented short-circuit parameter rapid evaluation method, system and equipment and a medium, and belongs to the technical field of power systems and automation, and the method comprises the steps: collecting and preprocessing operation data of protection configuration, and building a basic data set; initializing the short-circuit parameter evaluation model, carrying out initial evaluation on short-circuit parameters and analyzing the matching degree of the parameters; rapid evaluation of short-circuit parameters is carried out by using an iterative algorithm; based on the evaluation result, protection configuration optimization is carried out, and a dynamic constraint updating mechanism is set; performing adaptive adjustment and updating on the model parameters according to the operation data; and evaluating the evaluation precision by adopting multi-dimensional evaluation, and verifying the evaluation effect. According to the method, a complete technical closed loop including data preparation, rapid calculation, intelligent optimization, adaptive learning and effect verification is formed, the outstanding advantages of being high in evaluation speed, high in configuration precision and high in adaptive capacity are achieved, and a solution is provided for solving key problems in power system protection configuration.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of power systems and automation, in particular to a short-circuit parameter rapid evaluation method, system, device and medium for protection configuration. BACKGROUND

[0002] With the continuous expansion of the scale of power systems and the increasing complexity of network structure, accurate evaluation of short-circuit parameters is crucial for the rationality of protection configuration.

[0003] Traditional short-circuit parameter evaluation methods mainly rely on complex network calculations and a large number of simulation analyses, which are time-consuming and difficult to meet the needs of rapid protection configuration in engineering practice. In addition, existing evaluation methods often ignore the coupling relationship between protection device characteristics and short-circuit parameters, resulting in unsatisfactory protection configuration results. SUMMARY

[0004] In view of the above problems, the present application is proposed.

[0005] Therefore, the present application aims to solve the problems of complex calculation, long time consumption and low precision in short-circuit parameter evaluation. The present application proposes a short-circuit parameter rapid evaluation method for protection configuration, which establishes a mapping relationship between protection characteristics and short-circuit parameters to achieve rapid parameter evaluation and optimized protection configuration.

[0006] To solve the above technical problems, the present application provides the following technical solutions: a short-circuit parameter rapid evaluation method for protection configuration, which includes,

[0007] Collecting and preprocessing the operation data of protection configuration to establish a basic data set; initializing the short-circuit parameter evaluation model, performing initial evaluation of short-circuit parameters and analyzing the matching degree of parameters; using an iterative algorithm to perform rapid evaluation of short-circuit parameters; based on the evaluation results, performing protection configuration optimization and setting a dynamic constraint update mechanism; according to the operation data, adaptively adjusting and updating the model parameters; using multi-dimensional evaluation to measure the evaluation accuracy and verify the evaluation effect.

[0008] As a preferred solution of the short-circuit parameter rapid evaluation method for protection configuration, the preprocessing includes collecting operation data and defining an operation state vector.

[0009] The preprocessing is performed by standardization, and a data index structure is established to establish a basic data set.

[0010] As a preferred solution of the short-circuit parameter rapid evaluation method for protection configuration, the initialization of the short-circuit parameter evaluation model includes constructing a short-circuit parameter evaluation model suitable for protection configuration to perform initial evaluation of short-circuit parameters.

[0011] Analyzing the protection action characteristics, and establishing a protection characteristic model;

[0012] According to the matching degree evaluation method, the matching relationship between the protection characteristics and the system parameters is evaluated.

[0013] The beneficial effects of the preferred technical solutions in the embodiments of the present application are: the initial evaluation provides a reliable starting point, avoids starting from zero to search, improves the overall efficiency, and most importantly, through the analysis of the matching degree, the discordance between the existing protection configuration and the system demand can be accurately located, providing a clear target and basis for subsequent iterative optimization, so that the optimization process is targeted.

[0014] As a preferred scheme of the short-circuit parameter fast evaluation method for protection configuration according to the present application, wherein: the fast evaluation includes fast evaluation of short-circuit parameters through an improved iterative algorithm;

[0015] An optimization iteration strategy and a convergence criterion are set to determine whether the iteration is terminated.

[0016] The beneficial effects of the preferred technical solutions in the embodiments of the present application are: under the premise of ensuring the calculation accuracy, the evaluation time is significantly shortened. The optimization iteration strategy avoids invalid calculation, and the convergence criterion ensures that the algorithm is terminated immediately after the required accuracy is reached, preventing the waste of computing resources.

[0017] As a preferred scheme of the short-circuit parameter fast evaluation method for protection configuration according to the present application, wherein: the protection configuration optimization and the setting of the dynamic constraint update mechanism include: constructing a systematic protection configuration optimization method, establishing a mathematical model, and comprehensively considering system operation requirements, protection coordination principles and engineering actual constraints to set the optimal configuration of protection parameters;

[0018] Based on the multi-objective optimization theory, the optimization model of the protection configuration is constructed as:

[0019]

[0020] wherein, w i is the weight coefficient of the i-th protection point, h is the total number of protection points, M i is the corresponding characteristic matching degree, μ is the balance factor, c j is the j-th constraint condition, l is the total number of constraint conditions, and j is the variable index;

[0021] The dynamic constraint update mechanism is introduced to improve the optimization effect: wherein, ρ is the constraint update coefficient, Δc j is the change amount of the constraint condition, is the updated constraint condition, For the constraint condition before updating.

[0022] The beneficial effects of the preferred technical solutions in the embodiments of the application are that, through the multi-objective optimization model, the optimal solution or satisfactory solution meeting multiple competing requirements is found, and the one-sidedness and suboptimality of manual adjustment are avoided.

[0023] As a preferred scheme of the short-circuit parameter fast evaluation method for protection configuration, the adaptive adjustment and updating of the model parameters include proposing a gradient-based parameter adaptive adjustment method, and dynamically updating the model parameters by monitoring the system state change in real time.

[0024] Based on the deep learning theory, a parameter adaptive adjustment strategy is designed. Wherein, P t+1 and P t are the parameter values after updating and the current parameter values respectively; η is the learning rate of parameter updating is the gradient of the loss function with respect to the parameter;

[0025] At the same time, a dynamic learning rate adjustment mechanism is introduced: η t = η0(1+σt) -α Wherein, η0 is the initial learning rate, σ is the decay rate adjustment coefficient, t is the iteration number, and α is the decay index.

[0026] The beneficial effects of the preferred technical solutions in the embodiments of the application are that the evaluation model is given the ability of continuous learning and self-improvement, and with the continuous accumulation of operation data, the model will become more and more accurate, effectively overcoming the model drift problem caused by slow changing factors such as equipment aging and load growth.

[0027] As a preferred scheme of the short-circuit parameter fast evaluation method for protection configuration, the verification and evaluation effect includes constructing a multi-dimensional evaluation index system and introducing a root mean square error index to measure the evaluation accuracy.

[0028]

[0029] Wherein, Z a is the actual short-circuit impedance value of the a-th sample; is the corresponding evaluation value; A is the total number of samples, and a is the variable index;

[0030] At the same time, the relative accuracy index G is defined to verify the evaluation effect:

[0031]

[0032] wherein, e a is the evaluation error of the a-th sample, i.e. |Z a | is the absolute value of the actual impedance value; the smaller the value of E is, the higher the evaluation accuracy is, and the greater the value of G is, the better the evaluation effect is.

[0033] The preferred technical solutions in the embodiments of the application have the beneficial effects that the multi-dimensional indexes avoid the limitations of single indexes, can comprehensively reflect the evaluation accuracy and effect from different aspects, are not only the key to verify the effectiveness of the method, but also provide standardized and quantitative judgment basis for evaluating the performance of the method in actual engineering, and enhance the credibility and persuasiveness of the technical solutions.

[0034] Another object of the application is to provide a short-circuit parameter fast evaluation system oriented to protection configuration.

[0035] To solve the above technical problems, the application provides the following technical solutions: a short-circuit parameter fast evaluation system oriented to protection configuration, comprising: a data preprocessing module, a model initialization module, a fast evaluation module, a protection configuration optimization module, a parameter adjustment optimization module, and a performance evaluation verification module.

[0036] The data preprocessing module performs standardized processing on the operation data, establishes a basic data set for data cleaning and anomaly detection method, and simultaneously establishes a data index structure.

[0037] The model initialization module initializes the short-circuit parameter evaluation model, sets the initial value of the model parameter, constructs an evaluation matrix, and uses an improved parameter initialization method for model convergence.

[0038] The fast evaluation module uses an improved Newton iteration algorithm for fast evaluation of the short-circuit parameter, and performs dynamic step adjustment and convergence criterion optimization, simultaneously introduces parallel computing technology to speed up the evaluation process.

[0039] The protection configuration optimization module performs protection configuration optimization based on the evaluation result, balances the protection sensitivity and selectivity through a multi-objective optimization algorithm, and sets a dynamic constraint update mechanism.

[0040] The parameter adjustment optimization module adaptively adjusts the model parameter according to real-time operation data, and updates the parameter through gradient descent method.

[0041] The performance evaluation verification module uses multi-dimensional evaluation indexes to verify the effectiveness of the method, and tests the applicability and reliability of the algorithm through actual examples.

[0042] The application provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and wherein the processor implements the steps of the short-circuit parameter fast evaluation method for protection configuration when executing the computer program.

[0043] The application provides a computer readable storage medium, which stores a computer program, wherein the computer program implements the steps of the short-circuit parameter fast evaluation method for protection configuration when executed by a processor.

[0044] The application has the following beneficial effects: the short-circuit parameter fast evaluation method for protection configuration provided by the application realizes efficient and accurate evaluation of short-circuit parameters through a complete technical closed loop from data preprocessing, model initialization, fast iterative evaluation to dynamic optimization and adaptive updating. The method not only significantly improves the calculation speed by using an improved algorithm, but also ensures that the protection configuration scheme can automatically adapt to system changes and always remain in an optimal state through a multi-objective optimization model and a dynamic constraint updating mechanism. In addition, the parameter adaptive adjustment strategy based on the gradient enables the evaluation model to have a continuous learning ability, effectively overcoming the model drift problem. Finally, the reliability of the evaluation result is ensured through multi-dimensional verification indexes, thereby solving the time-consuming, rigid and insufficient precision problems of traditional methods, and greatly improving the safety and reliability of the power system. BRIEF DESCRIPTION OF DRAWINGS

[0045] In order to more clearly illustrate the technical solutions of the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor.

[0046] Figure 1 A general flowchart of the short-circuit parameter fast evaluation method for protection configuration provided by an embodiment of the application. DETAILED DESCRIPTION

[0047] In order to make the above-mentioned purposes, features and advantages of the application more obvious and easy to understand, the specific embodiments of the application will be described in detail below with reference to the drawings of the specification. Obviously, the described embodiments are only a part of the embodiments of the application, rather than all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor should be within the protection scope of the application.

[0048] Embodiment 1, refer to Figure 1 For an embodiment of the application, the embodiment provides a short-circuit parameter fast evaluation method for protection configuration, comprising:

[0049] S100, collect and pre-process the operation data of the protection configuration, and establish a basic data set;

[0050] S200, initialize the short-circuit parameter evaluation model, perform initial evaluation of the short-circuit parameters, and analyze the matching degree of the parameters;

[0051] S300, use an iterative algorithm to perform rapid evaluation of the short-circuit parameters;

[0052] S400, based on the evaluation results, perform protection configuration optimization and set a dynamic constraint update mechanism;

[0053] S500, according to the operation data, adaptively adjust and update the model parameters;

[0054] S600, use multi-dimensional evaluation to measure the evaluation accuracy and verify the evaluation effect.

[0055] It should be noted that the traditional short-circuit parameter evaluation method mainly relies on complex network calculation and a large amount of simulation analysis, which is time-consuming and difficult to meet the demand of rapid protection configuration in engineering practice. In addition, the existing evaluation method often ignores the coupling relationship between the protection device characteristics and the short-circuit parameters, resulting in unsatisfactory protection configuration effect.

[0056] Therefore, in view of the above problems, through the steps of S100-S600, based on the protection characteristic analysis theory, the correlation model of short-circuit parameters and protection configuration is constructed, the parameter calculation is realized through the rapid evaluation algorithm, and the protection configuration is performed combined with the optimization method.

[0057] Embodiment 2, refer to Figure 1 For an embodiment of the present application, the embodiment provides a short-circuit parameter rapid evaluation method for protection configuration, comprising:

[0058] In the embodiment of the present application, the operation data of the protection configuration is collected and pre-processed in S100, and a basic data set is established, including the following steps S101-S102:

[0059] S101, to realize rapid evaluation of short-circuit parameters, a complete system operation state description model is established, and based on the physical characteristics of the power system, the system operation state vector is defined as:

[0060] S={Z s ,I f ,V f ,R p ,Xp}

[0061] Wherein, Z s is the equivalent impedance of the system, reflecting the impedance characteristics of the system; I fFor fault current, the current size of the fault point is characterized; V f For fault voltage, the voltage level of the fault point is described; R p And X p The real part and the imaginary part of the protection impedance, respectively, are used to represent the impedance characteristics of the protection device.

[0062] In an embodiment of the present application, S102, preprocessing is performed to establish a basic data set, including the following steps A1-A3:

[0063] A1, considering the large difference in the dimension and scale of system parameters, a standardization processing method is introduced:

[0064]

[0065] Wherein, S is the original parameter value to be standardized, S min And S max The minimum and maximum values of the parameter in the historical data; S norm The standardized parameter value.

[0066] A2, data cleaning and anomaly detection method is used to ensure data quality.

[0067] A3, after standardization processing, all parameters are mapped to the interval [0, 1], which effectively eliminates the dimension effect.

[0068] In an optional embodiment, the standardization processing of S102 is to use a similar process as the original scheme, use statistical methods or box plot analysis to identify and process missing values and obvious abnormal records in the data, calculate the average value and standard deviation of each type of parameter data in the historical data set, and calculate the original data value of each parameter using the mean and standard deviation of the category it belongs to. The Z-Score value is converted, and all parameter data after standardization conversion is integrated to form a basic data set.

[0069] In another optional embodiment, the standardization processing of S102 also includes data cleaning, eliminating or correcting invalid and abnormal data, traversing each type of parameter, finding the maximum absolute value in the original data of the parameter, and determining that the maximum number is a power of 10. Divide each original data value under this type of parameter by a power of 10, so that the absolute value of all data is mapped to the interval [-1, 1], and most data will fall outside [-0.1, 0.1]. After scaling all parameters, a unified basic data set is formed.

[0070] In an embodiment of the present application, the short-circuit parameter evaluation model in S200 is initialized, the initial evaluation of the short-circuit parameters is performed, and the matching degree of the parameters is analyzed, including the following steps S201-S203:

[0071] S201, based on system operation characteristics and network topology, a short-circuit parameter evaluation model suitable for protection configuration is constructed, fully considering system impedance characteristics, network structure influence and weight distribution of various factors, realizing accurate evaluation of short-circuit parameters, and the evaluation model is represented as:

[0072] Z x =f(Z b ,K t ,α)+j·g(X b ,K x ,β)

[0073] Wherein, Z x is the short-circuit impedance to be evaluated, Z b and X b are the real and imaginary parts of the system reference impedance, K t and K x are the correction coefficients considering the influence of network structure, α and β are the weight factors reflecting the contribution of each part, f and g are the mapping functions of the real and imaginary parts respectively, and j is the imaginary unit.

[0074] In order to evaluate the accuracy of the model, the weighted root mean square error index ε is introduced:

[0075]

[0076] Wherein, Z i is the actual short-circuit impedance value of the i-th sample, is the corresponding evaluation value, w i is the weight coefficient reflecting the importance of the sample, n is the total number of samples, and i is the variable index; This index considers the evaluation accuracy and sample weight.

[0077] S202, in order to ensure that the short-circuit parameter evaluation results can effectively guide the protection configuration, the protection action characteristics are analyzed in depth, and a protection characteristic model F is established, which comprehensively considers three key dimensions of impedance characteristics, current characteristics and time characteristics, and realizes accurate description of the protection action process.

[0078] Based on modern protection theory and engineering practice experience, the protection action characteristic equation is:

[0079] F(Z,I,t)=k1Z+k2I+k3t-C≤0

[0080] Wherein, k1 is the impedance characteristic coefficient, reflecting the influence of impedance on protection action; k2 is the current characteristic coefficient, representing the contribution of current to protection action; k3 is the time characteristic coefficient, controlling the time characteristic of protection action; t is the protection action time; C is the protection action threshold value, determining the sensitivity of protection; Z is the impedance characteristic, and I is the current characteristic.

[0081] When the characteristic equation satisfies the inequality condition, the protection device acts. This characteristic analysis method combines short-circuit parameter evaluation and protection configuration organically, and provides a theoretical basis for subsequent protection optimization. Through the synergy of the above model, unified processing of short-circuit parameter evaluation and protection configuration is realized.

[0082] S203, the matching degree of the protection characteristic and the system parameter directly affects the reliability and sensitivity of the protection device, therefore, a matching degree evaluation method based on exponential weight is proposed, the matching relationship between the protection characteristic and the system parameter is accurately evaluated through quantitative calculation, and a scientific basis is provided for protection setting.

[0083] Based on in-depth theoretical analysis and engineering practice, a matching degree M evaluation function is designed:

[0084]

[0085] Wherein, ω i is the weight coefficient of different protection characteristics, reflecting the importance of each characteristic; λ is a characteristic attenuation factor, controlling the attenuation rate of the matching degree with the increase of the deviation; d i is the normalized deviation of each characteristic; m is the number of characteristic parameters.

[0086] In order to accurately quantify the characteristic deviation, a relative deviation calculation method is introduced:

[0087]

[0088] Wherein, Z i is the actual impedance value of the current evaluation; Z ref is the reference impedance value under standard working condition, used for normalization processing. This matching degree calculation method can effectively reflect the adaptation degree of the protection characteristic and the system parameter.

[0089] In the embodiment of the application, the short-circuit parameter is quickly evaluated by using an iterative algorithm in S300, including the following steps S301-S302:

[0090] S301, considering the requirement of calculation efficiency in engineering practice, a fast evaluation algorithm based on improved Newton iteration method is proposed, which significantly improves the calculation efficiency by optimizing the iteration strategy and convergence criterion.

[0091] Based on in-depth analysis of the evaluation problem, the improved Newton iteration algorithm is:

[0092]

[0093] Wherein, θ k is the parameter vector of the kth iteration, containing the short-circuit parameters to be evaluated; Hessian matrix of the objective function, reflecting the direction of parameter update; γ k adaptive step factor, controlling the amplitude of parameter update; gradient of the objective function.

[0094] In an embodiment of the present application, S302, setting the convergence criterion, comprises the following step B1:

[0095] B1, to ensure the convergence of the algorithm, set a strict convergence criterion:

[0096] ||θ k+1 -θ k ||≤δ

[0097] Where δ is a preset convergence threshold value for judging whether the iteration is terminated. The algorithm realizes the rapid and accurate evaluation of the parameters by dynamically adjusting the step size and accurate convergence judgment.

[0098] In an optional embodiment, S302 sets the convergence criterion as follows: after each iteration is completed, the absolute difference between the current objective function value and the last iteration objective function value is calculated. If the absolute difference is less than the preset convergence threshold value, it is determined that the convergence is achieved and the iteration is terminated. If the convergence is not achieved, the next round of iteration is continued.

[0099] In another optional embodiment, S302 sets the convergence criterion as follows: in each iteration, the norm of the gradient vector of the objective function is calculated. If the norm is less than the preset convergence threshold value, it is considered that the gradient is small enough and the extremum point is reached, and the iteration is terminated. If the condition is not met, the iteration is continued.

[0100] In an embodiment of the present application, S400, based on the evaluation result, the protection configuration optimization and the dynamic constraint update mechanism are set, comprising the following steps S401-S402:

[0101] S401, a systematic protection configuration optimization method is constructed. By establishing a mathematical model, the system operation requirements, protection coordination principles and engineering actual constraints are comprehensively considered, and the optimal configuration of the protection parameters is set.

[0102] Based on the multi-objective optimization theory, the optimization model of the protection configuration is constructed as follows:

[0103]

[0104] Where w i is the weight coefficient of the i th protection point, h is the total number of protection points, M i is the corresponding characteristic matching degree, μ is the balance factor, c j is the j th constraint condition, l is the total number of constraint conditions, and j is the variable index.

[0105] S402, introduce dynamic constraint update mechanism to improve optimization effect:

[0106]

[0107] Wherein, ρ is the constraint update coefficient, Δc j is the change of constraint condition, is the updated constraint condition, is the constraint condition before updating.

[0108] In the embodiment of the application, in S500, the model parameters are adaptively adjusted and updated according to the operation data, including the following steps S501-S502:

[0109] In the embodiment of the application, S501, the model parameters are adaptively adjusted, including the following steps C1-C2:

[0110] C1, the operation state of the power system has the characteristics of dynamic change, in order to ensure the accuracy of short-circuit parameter evaluation, a parameter adaptive adjustment method based on gradient is proposed, by real-time monitoring of system state change, dynamic updating of model parameters, realizing the rapid adaptation of evaluation model to system change.

[0111] C2, based on the theory of deep learning, a parameter adaptive adjustment strategy is designed:

[0112]

[0113] Wherein, P t+1 and P t are the updated and current parameter values respectively; η is the learning rate of parameter update, controlling the step length of parameter adjustment; is the gradient of the loss function with respect to the parameter, guiding the direction of parameter update.

[0114] In an optional embodiment, the adaptive adjustment of S501 is to continuously collect the latest operation data of the system, and to set a fixed length of recent data window as a reference set, the old data is removed from the window with the new data, ensuring that the reference set can reflect the recent state of the system; the current model parameters are applied to the recent data reference set for batch short-circuit parameter evaluation, and the average deviation between the evaluation results and the true value or expected value in the reference set is calculated; according to the calculated average deviation direction, the existing model parameters are gradually fine-tuned by using the moving average algorithm. The new parameter value is the weighted average of the old parameter and a fine-tuning value calculated based on the deviation, and the model parameters are continuously fine-tuned by using the new data window, so that the model slowly and stably tracks the long-term slow change of the system characteristics.

[0115] In another optional embodiment, the adaptive adjustment of S501 also sets one or more key monitoring indicators reflecting the system state change, and continuously monitors the values of these indicators, sets a threshold for triggering adjustment for each monitoring indicator. When the system is running normally, the change of the monitoring indicator does not exceed the threshold, it is determined that the system state is stable, the model parameters remain unchanged, and once the change of any one key indicator exceeds its preset threshold, it is determined that the system operating state has changed significantly. At this time, the adjustment mechanism is triggered, and the model parameters are reset to the initial values trained based on the current latest network topology and running data. After the parameter reset or switching, the system resumes monitoring of the key indicators, and waits for the next trigger condition.

[0116] S502, to improve the stability and adaptability of parameter adjustment, a dynamic learning rate adjustment mechanism is introduced:

[0117] η t =η0(1+σt) -α

[0118] Where η0 is the initial learning rate; σ is the decay rate adjustment coefficient, which controls the decay rate of the learning rate with time; t is the iteration number; α is the decay index, which is used to adjust the nonlinear variation characteristics of the learning rate.

[0119] This adaptive adjustment strategy can dynamically update the model parameters according to the change of the system state, improve the accuracy and adaptability of the evaluation.

[0120] In the embodiments of the present application, S600 adopts multi-dimensional evaluation to measure the evaluation accuracy and verify the evaluation effect, including the following steps S601-S602:

[0121] S601, to comprehensively evaluate the performance of the short circuit parameter evaluation method, a multi-dimensional evaluation index system is constructed, and a root mean square error index is introduced to measure the evaluation accuracy:

[0122]

[0123] Where Z a is the actual short circuit impedance value of the a-th sample; is the corresponding evaluation value; A is the total number of samples, and a is the variable index. This index reflects the overall accuracy of the evaluation result.

[0124] In the embodiments of the present application, S602, the evaluation effect is verified, including the following steps D1-D2:

[0125] D1, in order to more intuitively represent the evaluation performance, a relative accuracy rate G is defined:

[0126]

[0127] wherein e a is the evaluation error of the a-th sample, i.e. |Z a is the absolute value of the actual impedance value. This index converts the evaluation performance into a percentage form, facilitating performance judgment in engineering applications. These two evaluation indexes complement each other and jointly constitute the quantitative standard for evaluating the performance of the method.

[0128] The smaller the values of D2 and E are, the higher the evaluation accuracy is, and the larger the value of G is, the better the evaluation effect is. Through this evaluation system, the practical value of the patent method can be objectively evaluated, providing reliable performance guarantee for engineering applications.

[0129] In an optional embodiment, the verification and evaluation effect of S602 is to monitor the system operation state data in real time, calculate the deviation of the current parameter and the historical parameter; use the exponential weighted moving average (EWMA) method to update the model parameters smoothly, and the new parameter value is a linear combination of the historical weighted average value and the current value; dynamically adjust the weighting coefficient according to the change amplitude of the system state, increase the weight of the current value when the change is large, and increase the weight of the historical value when the change is small; periodically write the updated parameters into the evaluation model to complete the adaptive adjustment of the parameters.

[0130] In another optional embodiment, the verification and evaluation effect of S501 is also to set a sliding window with a fixed time length, continuously collect the system operation data within the window; calculate the average value of each parameter in the window as a reference to determine whether the current parameter deviates from the reference by more than a set threshold; if the deviation exceeds the threshold, adjust the model parameters to the weighted average result of the average value in the window and the current value; slide the window every certain time, recalculate the average value and update the parameters to realize gradual adjustment of the parameters.

[0131] Embodiment 3, as an embodiment of the present application, provides a short-circuit parameter fast evaluation method for protection configuration. In order to verify the beneficial effects of the present application, scientific demonstration is carried out through experiments.

[0132] MATLAB R2023a is used for algorithm verification, and simulation test is carried out based on IEEE 39-node standard system. The test system includes 39 bus nodes, 10 generators and 46 transmission lines.

[0133] As shown in Table 1, the performance comparison test selects four typical methods: traditional method (based on impedance ratio calculation), improved method (introducing adaptive factor), optimization method (based on gradient descent) and the method proposed in this paper. The evaluation indexes include root mean square error (RMSE), evaluation time, accuracy and memory occupation. Among them, RMSE reflects the deviation degree of the evaluation value and the true value, the evaluation time represents the real-time performance of the algorithm, the accuracy statistics the correct proportion of the evaluation result, and the memory occupation reflects the resource demand of the algorithm.

[0134] Evaluation performance comparison of different methods

[0135] Evaluation method RMSE(Ω) Evaluation time (ms) Accuracy rate (%) Memory usage (MB) Traditional method 0.256 458 85.6 456 Improved method 0.198 325 88.4 389 Optimized method 0.156 246 92.8 312 The method 0.085 168 95.3 245

[0136] The evaluation effects for different types of faults are shown in Table 2. Four typical fault scenarios are designed: single-phase ground fault (the most common fault type, with a ground resistance of 0.1 Ω), two-phase short-circuit fault (with a short-circuit impedance of 0.01 Ω), three-phase short-circuit fault (metallic short-circuit), and phase-to-phase short-circuit fault (with a short-circuit point resistance of 0.05 Ω). Each fault type is tested 50 times, with faults triggered randomly at different locations and times. Accuracy reflects the correctness of the evaluation results, response time represents the time interval from fault occurrence to evaluation result, and matching degree describes the degree of agreement between the evaluation result and the actual parameters.

[0137] Evaluation effects under different fault types

[0138] Fault type Accuracy rate (%) Response time (ms) Matching degree (%) Single-phase ground fault 97.2 165 96.5 Two-phase short circuit 94.5 172 93.8 Three-phase short circuit 93.8 178 92.6 Inter-phase short circuit 92.6 185 91.4

[0139] System performance evaluation, as shown in Table 3, focuses on four key indicators: model stability reflects the reliability of the algorithm in long-term operation, computational efficiency represents CPU utilization and task processing capability, memory utilization monitors system resource occupation, and evaluation accuracy counts the accuracy of the overall evaluation results. During the test, the system runs continuously for 72 hours, and complete performance indicators are recorded every hour.

[0140] System performance evaluation indicators

[0141] Evaluation index Numerical value Model stability (%) 99.2 Computing efficiency (%) 96.5 Memory utilization rate (%) 45.6 Evaluation accuracy (%) 95.3

[0142] The experimental results show that the proposed method is superior to the traditional method in terms of evaluation accuracy and computational efficiency, with RMSE reduced to 0.085 Ω, evaluation time shortened to 168 ms, and accuracy improved to 95.3%. The method shows good adaptability under various fault conditions and meets the practical engineering requirements. The entire evaluation process is computationally efficient and reliable, with good engineering application value.

[0143] In one embodiment of the present application, the above is a schematic scheme of a short-circuit parameter fast evaluation method for protection configuration. It should be noted that the technical scheme of a short-circuit parameter fast evaluation system for protection configuration and the technical scheme of the above short-circuit parameter fast evaluation method for protection configuration belong to the same concept. The technical scheme of the short-circuit parameter fast evaluation system for protection configuration in this embodiment, which is not described in detail, can be referred to the description of the technical scheme of the short-circuit parameter fast evaluation method for protection configuration.

[0144] The embodiment provides a short-circuit parameter fast evaluation system for protection configuration, which comprises a data preprocessing module, a model initialization module, a fast evaluation module, a protection configuration optimization module, a parameter adjustment optimization module and a performance evaluation verification module.

[0145] The data preprocessing module performs standardization processing on operation data, establishes a basic data set, adopts a data cleaning and abnormality detection method to ensure data quality, and establishes a data index structure to improve query efficiency.

[0146] The model initialization module initializes a short-circuit parameter evaluation model based on a system topological structure, sets initial values of model parameters, constructs an evaluation matrix, and adopts an improved parameter initialization method to improve model convergence.

[0147] The fast evaluation module uses an improved Newton iteration algorithm to realize fast evaluation of short-circuit parameters, improves algorithm efficiency through dynamic step adjustment and convergence criterion optimization, and introduces parallel computing technology to speed up the evaluation process.

[0148] The protection configuration optimization module performs protection configuration optimization based on evaluation results, balances protection sensitivity and selectivity through a multi-objective optimization algorithm, and adopts a dynamic constraint update mechanism to ensure the feasibility of optimization results.

[0149] The parameter adjustment optimization module adaptively adjusts model parameters according to real-time operation data, updates parameters through a gradient descent method to realize online optimization of the model, and introduces an early stopping mechanism to prevent overfitting.

[0150] The performance evaluation verification module uses multi-dimensional evaluation indexes to verify the effectiveness of the method, and verifies the applicability and reliability of the algorithm through actual example tests.

[0151] The embodiment also provides an electronic device suitable for a short-circuit parameter fast evaluation method for protection configuration, which comprises a memory and a processor.

[0152] The embodiment also provides a storage medium having a computer program stored thereon, and the program is executed by a processor to realize a short-circuit parameter fast evaluation method for protection configuration.

[0153] The storage medium provided by the embodiment and the short-circuit parameter fast evaluation method for protection configuration provided by the above embodiment belong to the same inventive concept, and the technical details not described in the embodiment can be referred to the above embodiment, and the embodiment has the same beneficial effects as the above embodiment.

[0154] From the above description of the embodiments, those skilled in the art can clearly understand that the present application can be implemented by means of software and necessary universal hardware, and of course can also be implemented by hardware, but in many cases the former is a better implementation. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, and the computer software product can be stored in a computer readable storage medium, such as a floppy disk, a read-only memory (ROM), a random access memory (RAM), a FLASH memory, a hard disk, or an optical disc, etc., and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods of various embodiments of the present application.

[0155] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application but not limit the present application, and although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the present application, and all of them should be covered in the scope of the claims of the present application.

Claims

1. A short circuit parameter fast evaluation method oriented to protection configuration, characterized in that: The application relates to a short-circuit parameter evaluation method for power system protection configuration, which comprises the following steps: Collecting and preprocessing operation data of protection configuration to establish a basic data set; Initializing a short-circuit parameter evaluation model, performing initial evaluation of the short-circuit parameter and analyzing matching degree of the parameter; Performing rapid evaluation of the short-circuit parameter by using an iterative algorithm; Based on the evaluation result, performing protection configuration optimization and setting a dynamic constraint updating mechanism; According to the operation data, performing adaptive adjustment and updating of model parameters; Adopting multi-dimensional evaluation to measure evaluation accuracy and verify evaluation effect.

2. The method for fast short circuit parameter evaluation for protection-oriented configuration according to claim 1, characterized in that: The preprocessing comprises collecting operation data and defining an operation state vector; Through standardization processing, the preprocessing is performed, meanwhile, a data index structure is established to establish the basic data set.

3. The method for fast short circuit parameter evaluation for protection-oriented configuration according to claim 2, characterized in that: The initializing of the short-circuit parameter evaluation model comprises constructing a short-circuit parameter evaluation model suitable for protection configuration to perform initial evaluation of the short-circuit parameter; Analyzing protection action characteristics to establish a protection characteristic model; According to a matching degree evaluation method, the matching relationship between protection characteristics and system parameters is evaluated.

4. The method of claim 3, wherein the short circuit parameter fast evaluation method is oriented to a protection configuration. The rapid evaluation comprises performing rapid evaluation of the short-circuit parameter by using an improved iterative algorithm; Setting an optimization iterative strategy and a convergence criterion to judge whether the iteration is terminated.

5. The method for fast short circuit parameter evaluation for protection oriented configuration according to claim 4, characterized in that: The protection configuration optimization and the setting of the dynamic constraint updating mechanism comprise constructing a systematic protection configuration optimization method, setting an optimal configuration of protection parameters by establishing a mathematical model and comprehensively considering system operation requirements, protection coordination principles and engineering actual constraints; Based on a multi-objective optimization theory, an optimization model of the protection configuration is constructed as follows: where w i is the weight coefficient of the i-th protection point, h is the total number of protection points, M i is the corresponding characteristic matching degree, μ is the balance factor, c j is the j-th constraint condition, l is the total number of constraint conditions, and j is the variable index; The dynamic constraint updating mechanism is introduced to improve the optimization effect: wherein, p is a constraint updating coefficient, and Δc j is a change of the constraint condition, is an updated constraint condition, is a constraint condition before updating.

6. The method for fast short circuit parameter evaluation for protection-oriented configuration according to claim 5, characterized in that: The adaptive adjustment and updating of the model parameters comprise proposing a gradient-based parameter adaptive adjustment method, and dynamically updating model parameters by monitoring system state changes in real time; Based on the deep learning theory, a parameter self-adaptive adjustment strategy is designed: wherein, P t+1 and P t are the updated and current parameter values respectively; η is the learning rate of parameter update is the gradient of the loss function with respect to the parameter; At the same time, a dynamic learning rate adjustment mechanism is introduced: η t = η0(1 + σt) -α where η0is the initial learning rate, σ is the decay rate adjustment coefficient, t is the iteration number, and a is the decay index.

7. The method for fast short circuit parameter evaluation for protection-oriented configuration according to claim 6, characterized in that: The verification of the evaluation effect comprises constructing a multi-dimensional evaluation index system and introducing a root mean square error index to measure evaluation accuracy: wherein Z a is the actual short-circuit impedance value of the a-th sample; is the corresponding evaluation value; A is the total number of samples, and a is the variable index. Meanwhile, a relative accuracy index G is defined to verify the evaluation effect: Wherein, e a is the evaluation error of the a-th sample, i.e. |Z a | is the absolute value of the actual impedance value; the smaller the E value, the higher the evaluation accuracy, and the larger the G value, the better the evaluation effect.

8. A short-circuit parameter fast evaluation system for protection configuration, applying the short-circuit parameter fast evaluation method for protection configuration as claimed in any one of claims 1 to 7, characterized in that, The application further discloses a short-circuit parameter evaluation method for power system protection configuration. The application comprises the following modules: A data preprocessing module, a model initializing module, a rapid evaluation module, a protection configuration optimization module, a parameter adjustment optimization module and a performance evaluation verification module; The data preprocessing module performs standardization processing on operation data, establishes a basic data set, performs data cleaning and abnormality detection and simultaneously establishes a data index structure; The model initializing module initializes a short-circuit parameter evaluation model, sets initial values of model parameters, constructs an evaluation matrix and adopts an improved parameter initialization method to perform model convergence; The rapid evaluation module utilizes an improved Newton iterative algorithm to perform rapid evaluation of the short-circuit parameter, performs dynamic step adjustment and convergence criterion optimization, simultaneously introduces parallel computing technology to accelerate the evaluation process; The protection configuration optimization module performs protection configuration optimization based on the evaluation result, balances protection sensitivity and selectivity by using a multi-objective optimization algorithm and sets a dynamic constraint updating mechanism; The parameter adjustment optimization module performs adaptive adjustment of model parameters according to real-time operation data and updates the parameters by using a gradient descent method; The performance evaluation verification module adopts multi-dimensional evaluation indexes to verify effectiveness of the method and verifies applicability and reliability of the algorithm through actual example tests. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-8 when the computer program is executed by the processor. The computer program is executed by the processor to implement the steps of the short-circuit parameter fast evaluation method for protection configuration in any one of claims 1 to 7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the short-circuit parameter fast evaluation method for protection configuration in any one of claims 1 to 7.

Images